As operators deploy advanced techniques like Massive MIMO and network slicing, AI becomes increasingly essential for scalable performance management. This year, the AI-RAN Alliance welcomed Rimedo Labs, a company tightly focused on improving outcomes for operators and bridging the gap between cutting-edge academic research and the commercial challenges of the telecom industry.

To learn more about Rimedo Labs, we checked in with Technical Solution Manager Marcin Hoffmann, who leads R&D projects and the company’s focus on developing practical RAN automation solutions, particularly xApps and rApps to address key operator challenges in traffic steering, energy efficiency, security, and Massive MIMO optimization.

Bridging Research and Reality

“Rimedo Labs is essentially an R&D hub for the telecom industry, specializing in providing high-quality RAN algorithm design and implementation tailored to customers’ needs and requirements, applied research, and consulting services in the field of modern wireless systems,” explains Hoffmann. “As a university spin-off, we translate deep academic research, especially in AI/ML, into solutions for the practical, commercial needs of MNOs and vendors.”

The company’s mission is simple yet ambitious: to help operators achieve smarter, more autonomous networks that adapt dynamically to the ever-changing needs of users and services.

Moving from Theory to Practice with the AI-RAN Alliance

Rimedo Labs joined the AI-RAN Alliance to collaborate with leading mobile network operators (MNOs) and technology partners in transforming the way networks are managed. “We see a clear market demand for more dynamic, context-dependent management of mobile networks to guarantee a better user experience,” says Hoffmann. “We believe that this can be achieved by using AI. Within the AI-RAN Alliance, Rimedo Labs has an opportunity to interact with MNOs and other vendor partners to bring these advanced AI concepts from theory into a commercial reality.”

AI: Changing the RAN Equation

Rimedo and its R&D team recognize the unique complexity of RAN environments, integrating diverse service types -each with distinct QoS requirements. Factor in emerging technologies like Massive MIMO, and, from the Rimedo viewpoint, traditional optimization approaches have reached their limits.

AI changes the equation. By enabling data-based reasoning and continuous learning, AI automates optimization, unlocking new levels of performance, adaptability, and energy efficiency that would be impossible through analytical methods alone.

Aligning with the AI-RAN Vision

“Rimedo Labs aims to develop intelligent, AI-based algorithms to optimize RAN in terms of, e.g., traffic steering, energy savings, or security. While this mission started with the Open RAN concept, it fits perfectly with the AI-RAN Alliance, especially in the context of revolutionizing RAN optimization with AI,” says Hoffmann.

Forecasting A Shift

Rimedo Labs notes a clear demand among MNOs for intelligent autonomous RAN management and sees an opportunity for the AI-RAN Alliance to demonstrate how AI can fill this unmet need, shifting networks from reactive optimization to a predictive state that radically lowers OpEx and enables new high-value services through dynamic QoS. “Especially while the 6G standardization work is ongoing, the key opportunity for the AI-RAN Alliance is to emphasize this foundational, AI-native nature of 6G networks,” observes Hoffmann.

Engagement Aligned with Expertise

“The AI-for-RAN working group is a natural fit for Rimedo Labs, as it is focused directly on the optimization of RAN using AI,” summarizes Hoffmann. “Moreover, the use cases fit the Rimedo Labs expertise, e.g., traffic steering, energy savings, security, or beam management in Massive MIMO.”

“Apart from WG1,” Hoffman states, “Rimedo Labs is also highly interested in the D4AI initiative. We believe that to enable AI-based RAN optimization, there is a need for unified access to the live-network data. First, to understand the real-world problems. Second, to train the proper AI model to resolve it. Moreover, while Rimedo Labs has experience working with MNO network data, we promote a synthetic data generation approach to use a subset of live network data to generate more datasets, e.g., using Digital Twin or generative AI. This allows for testing AI-for-RAN algorithms under a variety of scenarios. Our team is particularly active in the AI-for-RAN Working Group, which focuses on use cases that align perfectly with our expertise-traffic steering, energy savings, security, and beam management in Massive MIMO.”

Rimedo is also involved in the D4AI initiative, promoting a unified approach to accessing live-network data and synthetic data generation using Digital Twins and generative AI. The company considers these tools to be essential for training robust AI models and testing them across diverse network scenarios.

Learn More About Membership

The AI-RAN Alliance thrives on collaboration across the telecom and AI communities. To learn how your organization can contribute to the future of AI-RAN, visit our membership page.